similar to: "a.matrix[a.char.matrix]<- " crashes R (PR#447)

Is this a bug? > matrix(1:4,nc=2,dimnames=list(1:2,1:2))[c(1,3)] 1 NA 1 3 It has some annoying consequences, e.g. > median( matrix(1:9,nc=3,dimnames=list(1:3,1:3))[1,,drop=F] ) NA 4 > The above was for Version: platform = sparc-sun-solaris2.7 arch = sparc os = solaris2.7 system = sparc, solaris2.7 status = major = 1 minor = 3.0 year = 2001 month = 06 day = 22

examples: 1) Fails to report that components 2 and 3 differ all.equal(list(1,2,3,zap=1),list(1,3,4,zap=2)) [1] "Component zap: Mean relative difference: 1 2) Incorrectly asserts all are equal when components 2 and 3 differ > all.equal(list(1,2,3,zap=1),list(1,3,4,zap=1)) [1] TRUE 3) Removing named component reveals differences: > all.equal(list(1,2,3,1),list(1,3,4,1)) [1]

> split(c(a=1,b=2),c(1,2)) $"1" [1] 1 $"2" [1] 2 The documentation does not *promise* that names will survive, but in Splus3.4, they do. This broke a function that worked under S+ that used the names attribute of elements in the components returned by split. Chuck Berry --please do not edit the information below-- Version: platform = sparc-sun-solaris2.7 arch = sparc

Problem: > names(my.dat) [1] "x" "y" > cor(my.dat$x,my.dat$y) # desired result [1] 0.8660254 > cor(my.dat$x,my.dat$y.NOT) # typo [1] 1 > I'd prefer a warning() or stop() in that case. Solution (??) : *** 1,5 **** --- 1,7 ---- cor <- function (x, y=NULL, use="all.obs") { + if (is.null(y) && !is.null( match.call()$y ) ) +

<<insert bug report here>> IMHO, plot(x,c(1)$nothing) should fail or at least issue a warning. Instead it produces an index plot of the first argument, i.e. plot(x) Chuck Berry --please do not edit the information below-- Version: platform = sparc-sun-solaris2.7 arch = sparc os = solaris2.7 system = sparc, solaris2.7 status = Patched major = 1 minor = 0.0 year = 2000

Since the Extract page has usage as: x[i, j, ... , drop=TRUE] I would expect that 'drop=' would need to be given to the third 'subscript' > diag(4)[ , 4 , 4 ] # Forgot 'drop=' or added extra ',' Error in diag(4)[, 4, 4] : incorrect number of dimensions > > as.data.frame( diag(4) )[ , 4 , 4 ] # should return error, right? [1] 0 0 0 1 Also note:

viz. > (function(x,y=a) substitute(x$y))(x) x$a > (function(x,y=a) substitute(x$y))(list(y=1:3)) list(y = 1:3)$a > (function(x,y) eval(substitute(substitute(y))))(x=list(y=1:3),y=x$y) list(y = 1:3)$x$y The behavior I expect and want is like that in Splus 3.4: > (function(x,y=a) substitute(x$y))(x) x$y > (function(x,y=a) substitute(x$y))(list(y=1:3)) list(y = 1:3)$y >

Dear R-help, I tried to build the R-release version of R-1.0.0. ./configure and make seemed to go just fine. When I ran 'make check' from $RHOME, it dumped core apparently in the quit('no') on the last line of base-Ex.R I had been having some problems with a package I am building in that R would report with "Process R segmentation violation (core dumped)" upon

<<insert bug report here>> > apply(array(1:20,c(2,2,5)),2:3,function(x) x) Error: length of dimnames must match that of dims > Changing: dimnames = if (is.null(dn.ans)) list(ans.names, NULL) else c(list(ans.names), dn.ans) To: dimnames = if (length(dn)==0) NULL else if (is.null(dn.ans)) list(ans.names, NULL) else c(list(ans.names), dn.ans) seems to fix this. Chuck

I get a segmentation fault as follows: > tmp <- array(list(), c(3, 4)) > tmp[[1, 2]] NULL # as expected > tmp[[1, 2]]$fred <- 1:10 > tmp # as expected [,1] [,2] [,3] [,4] [1,] "NULL" "List,1" "NULL" "NULL" [2,] "NULL" "NULL"

Another one from the student computer practicals: it is possible to create a dataframe that cannot be printed. Thus > fred <- data.frame(happy=c(TRUE, FALSE, TRUE), sad=7:9) > tmp <- fred[c(FALSE, FALSE, TRUE, TRUE)] # no error message! > tmp Error in as.matrix.data.frame(x) : dim<- length of dims do not match the length of object > print.default(tmp) $"NA" NULL

Here's a nasty one. The following has caused a segmentation fault and possibly also a bus error. fred <- 1:6 dim(fred) <- 6 dimnames(fred) <- list(LETTERS[1:6]) mm <- matrix(1:12, 2, 6) mm %*% fred # segmentation fault here In the case without the dimnames assignment the result is OK. Cheers, Jonathan. --please do not edit the information below-- Version: platform =

This creates a named vector of length nx, then repeatedly draws a single sample from it. lkup <- function(nx, m=10000L) { tbl <- seq_len(nx) names(tbl) <- as.character(tbl) v <- sample(names(tbl), m, replace=TRUE) system.time(for(k in v) tbl[k], gcFirst=TRUE) } There is an abrupt performance degredation at nx=1000 > lkup(1000) user system elapsed 0.180

The following commands z <- matrix(seq(0, 1, len=1000)) image(z, 1, z, zlim=c(0, 1), col=gray(1:10/10)) appear to violate the image functionality as described on the help page. As I understand it, by specifying a zlim of c(0, 1) and a colour vector of length 10, the colours should correspond to the intervals (0, 0.1), (0.1, 0.2), ..., (0.9, 1.0), this being 10 equally-spaced intervals on (0,

There appears to be a mild bug, or at least a deficiency, in rnorm. The bug becomes apparent when one looks at extremes of the squares of the values generated by rnorm; rnorm is not generating quite enough extreme values. The R version that I am using is 1.4.1; I never got around to installing 1.5.0, and now since 1.5.1 is about to come out .... However, checking the 1.5.0 release notes

Has anyone seen this issue before on solaris 2.7. I keep getting this error for libshout when trying to compile the example. root#[/usr/local/src/libshout-1.0.9/example]#gcc -lshout -o test example.c Undefined first referenced symbol in file socket /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/3.2/../../../libshout.so recv

thanks Karl, that got me almost there. I think i am still missing one other library apparently. ANy ideas ? #gcc -lshout -lsocket -lnsl -o test example.c Undefined first referenced symbol in file inet_pton /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/3.2/../../../libshout.so ld: fatal: Symbol referencing errors. No output written to test

This question is about compiling R-1.6.2 from source on a Sparcstation-20 machine running Solaris 7, gcc-2.95.3 and readline-4.2 (both gcc and readline from sunfreeware.com). I searched the r-help archive and the FAQ. The errors seem to be related to the binary readline libraries, and I have /usr/local/lib/libreadline.a,libreadline.so@, and libreadline.so.4. The attached errors are from

Is there a way of `restarting' split.screen? This is what I am getting: > close.screen() [1] 10 11 12 13 > close.screen(all=TRUE) Error in par(args) : parameter "i" in "mfg" is out of range > graphics.off() > x11() > close.screen() [1] 10 11 12 13 > close.screen(all=TRUE) Error in par(args) : parameter "i" in "mfg" is out of range As

Package bqtl version 1.0 is now available on CRAN. Description: QTL mapping toolkit for inbred crosses and recombinant inbred lines. Includes maximum likelihood and Bayesian tools. I am keen to have comments about this package including implementation details, additional functionality, and (of course) problems and bugs. Chuck Berry -- Charles C. Berry (858) 534-2098